Osteoporosis, a degenerative bone disease characterized by reductions in bone volume and by increased risk of spontaneous fracture of the hip and spine, affects millions. Despite the fact that bone fracture represents the overriding clinical consequence of osteoporosis, the biomechanical aspects of the disease have received little attention. It is not currently possible, on the basis of available clinical assessments, to predict reliably which patients are at risk for the development of spontaneous fractures. Therefore, the choice to use therapeutic modalities, which are themselves associated with significant risks, is being made without benefit of accurate fracture predictors. The primary objective of the proposed investigation is to develop improved predictors of vertebral fracture risk based first on biomechanical models and controlled in-vitro strength testing of cadaver vertebrae and then on verification and refinement in a well defined patient population. Vertebral density and geometry from fresh cadaver lumbar spine segments will first be characterized in-vitro using radiographic and CT data validated against independent direct measurements of these parameters. The utimate strength of vertebral FSU will then be determined using two loading modes: 1) uniaxial compression; and 2) combined compression and flexion. Vertebral strength results will be compared against radiographic and CT data and against forearm densitometric and iliac crest histomorphometric evaluations from the same cadavers. A representative three-dimensional finite element model of the vertebral body will be generated to study parametrically the most important geometric and constitutive properties influencing vertebral strength. A clinical trial of the resulting biomechanical predictors of fracture strength will be conducted using available patient data from the MGH Bone Metabolism Unit, each with complete metabolic studies, quantitative CT scans, plane spine films and forearm absorption values. All patients will be followed during the course of the study for clinical evidence of spontaneous vertebral fracture and the various assessment parameters compared for their discriminatory capabilities. We hope thereby to identify those biomechanical parameters most critical to vertebral fracture. The eventual goal is to provide an accurate clinical tool which may be used to assess fracture risk and help guide clinical decisions to implement therapeutic modalities.